Addiction to alcohol and illicit drugs is the third leading cause of preventable death in the United States. It is well known that genetic factors play a major role in the development of addiction, and many of these factors appear to contribute to a generalized liability for addiction. However, environmental factors also play important roles, and there is abundant evidence that a number of environmental factors may alter the effect of genetic risk factors. Hence, it is critical to approach addictive disease using a gene-environment approach. We propose to undertake a comprehensive genome-wide association study using nearly 3,000 unrelated subjects of European descent in a case-control design and to analyze genetic associations and gene x environment interactions. Cases include 1,333 subjects with the primary phenotype of DSM-IV alcohol dependence, and often with other comorbid drug addiction. Controls comprise 1,647 subjects who have used alcohol, but have never been addicted to alcohol, nicotine, or other illicit substances. This approach to control status will ensure that genetic findings are related to substance dependence, rather than the initiation of substance use. The large number of subjects is made feasible by a cross-study collaboration involving three major studies of addictive disorders: the Collaborative Study on the Genetics of Alcoholism (COGA), the Family Study of Cocaine Dependence (FSCD), and the Collaborative Genetic Study of Nicotine Dependence (COGEND). All these studies participate in the NIAAA and NIDA Genetic Repositories. Our primary goals of this proposal are to undertake large-scale genotyping through the Genes and Environment Initiative (GEI) sequencing center, to standardize and harmonize phenotypic and environmental variables across all studies, and to perform a logical and sequential set of analyses that will incorporate covariates and environmental features for the testing of genetic association, gene-environment and gene-gene interactions. Finally, we will seek to replicate and extend findings from this genome-wide association study in an African-American population with over 1300 subjects (760 cases and 485 controls). Other independent and genetically informative datasets, including two data sets collected by COGA, and three large data sets held by collaborating investigators, provide the opportunity to further investigate genetic findings. This study will be the first to systematically address both genetic and environmental effects in addiction to using a genome-wide association approach;novel features include the integration of multiple large-scale data sets, a carefully considered definition of control status, and the opportunity for replication in several independent samples.